Recently, we corrected sickle cell disease in a mouse model by transducing purified, unmobilized hematopoietic stem cells (HSCs) with a lentiviral vector carrying an anti-sickling beta-globin gene (betaAS3). Unlike previously reported studies, HSCs were purified from unmobilized marrow to model a transduction protocol that may be translated directly to human patients. Also, 1,000 times less virus was used compared to previous studies in an attempt to develop a safer transduction protocol. We propose to utilize this same protocol to transduce purified, human hematopoietic stem cells from sickle cell patients. Although this approach is powerful, one weakness is that the site of lentiviral integration cannot be determined before cells are transplanted back into the recipient. An approach that bypasses this potential problem is to transduce embryonic stem cells. Viral integration sites can then be sequenced in multiple colonies to define cells that contain insertions in innocuous positions in the genome. These cells can then be differentiated into hematopoietic stem cells and subsequently transplanted into recipients. We have recently corrected our mouse model of beta-thalassemia by lentiviral transduction of ES cells and we propose to do similar experiments in our mouse model of sickle cell disease. Finally, we have recently produced a new mouse model of sickle cell disease by replacing the mouse beta- globin genes with the human betaS gene and replacing the mouse alpha-globin genes with human alpha-globin genes in ES cells. Mice derived from these knockin modifications develop all of the pathology of our previous knockout, transgenic mouse model. We have now derived ES cells from these mice and propose to correct the sickle gene by homologous recombination. This protocol avoids random insertion. These cells will then be differentiated into HSCs and transplanted into irradiated sickle mouse recipients. Although this approach cannot be translated to humans at the present time, the studies will test the efficacy and safety of this protocol. [unreadable] [unreadable] [unreadable]